Skip to Main Content
Data Mining provides a useful insight on finding frequent patterns. Data mining is an inter-disciplinary field, whose core is at the intersection of machine learning, statistics and databases. In this paper we proposed an efficient method for knowledge discovery which is based on subset and superset approach. In this approach we also use dynamic minimum support so that we reduce the execution time. A frequent superset means it contains more transactions then the minimum support. It utilize the concept that if the item set is not frequent but the superset may be frequent which is consider for the further data mining task. By this approach we can also find improved association, which shows that which item set is most acceptable association with others. A frequent subset means it contains less transactions then the minimum support. It utilizes the behavior that the less count may be frequent if we attached the less count with the higher order set. Here we also provide the flexibility to find multiple minimum supports which is useful for comparison with associated items and dynamic support range. Our algorithm provides the flexibility for improved association and dynamic support. Comparative result shows the effectiveness of our algorithm.